Bug#403031: proposed patch to clean the finish-install serial console detection.

[Sven Luther <sven.luther@wanadoo.fr>]

On Tue, Dec 19, 2006 at 03:21:11PM +0100, Frans Pop wrote:
> On Saturday 16 December 2006 00:30, Sven Luther wrote:
> > Please find attached a patch which does the finish-install detection of
> > a serial console, based on what was done in
> > rootskel/detect-linux-console.
> 
> Has this patch been tested?

Sure, i used it on the efika. It causes no regresion on the hardware i have,
but i agree with you that it is important to test on other hardware.

> Even if it has, I'm not sure we should make this change so short before 
> the release because of the risk of regressions. I'd rather just add the 
> support for the missing type of devices for now and leave this BR open so 
> this patch can be applied post-Etch.

Err, can you tell me the risk of regression. Please speak the issue with
someone else, like Colin and Joey, or Bastian, and weigth the actual risk.

The problem would be if there is a legitimate reason why a certain device
would be declared as serial console in rootskel, but not in finish-install, i
can't myself imagine where this could be, and when i did an audit for ttyS* in
the whole d-i packages svn repo, i didn't find any other place which is
involved, but this was a couple of months ago already.

So, that you apply it now or later, i would be very interested in the reason
why this was not done as this patch in the first place, which would seem more
logical and intuitive.

Friendly,

Sven Luther